Optical Coherence Tomography Angiography (OCTA) is a promising diagnostic tool for age-related macular degeneration (AMD), providing non-invasive visualization of sub-retinal vascular networks. This research explores the effectiveness of deep neural network (DNN) classifiers trained exclusively on OCTA images for AMD diagnosis. To address the challenge of limited data, we combine OCTA data from two instruments-Heidelberg and Optovue-and leverage style transfer technique, CycleGAN, to convert samples between these domains. This strategy introduces additional content into each domain, enriching the training dataset and improving classification accuracy. To enhance the CycleGAN for downstream classification tasks, we propose integrating class-related constraints during training, which can be implemented in either supervised or unsupervised manner with a pretrained classifier. The experimental results demonstrate that the proposed class-conditioned CycleGAN is effective and elevates DNN classification accuracy in both OCTA domains.

This study investigates the efficacy of predicting age-related macular degeneration (AMD) activity through deep neural networks (DNN) using a cross-instrument training dataset composed of Optical coherence tomography-angiography (OCTA) images from two different manufacturers. A retrospective cross-sectional study analyzed 2D vascular en-face OCTA images from Heidelberg Spectralis (1478 samples: 1102 training, 276 validation, 100 testing) and Optovue Solix (1003 samples: 754 training, 189 validation, 60 testing). OCTA scans were labeled based on clinical diagnoses and adjacent B-scan OCT fluid information, categorizing activity into normal, dry AMD, active wet AMD, and wet AMD in remission. Experiments explored cross-instrument disease classification using separate and combined datasets for training the DNN. Testing involved 100 Heidelberg and 60 Optovue samples. Training on Heidelberg data alone yielded 73% accuracy on Heidelberg images and 60% on Optovue images. Training on Optovue data alone resulted in 34% accuracy on Heidelberg and 85% on Optovue images. Combined training data from both instruments achieved 78% accuracy on Heidelberg and 76% on Optovue test sets. Results indicate that cross-instrument classifier training demonstrates high classification prediction accuracy, making cross-instrument training viable for future clinical applications. This implies that vascular morphology in OCTA can predict disease progression.